Schaeffer Center Paper is Among Most Downloaded in Journal of Mathematical Psychology
School of Pharmacy Professor Jason Doctor is a co-author on the 2011 paper that ranks as the 10th most downloaded paper in the journal’s history.
School of Pharmacy associate professor Jason Doctor is a co-author on a 2011 paper that is the 10th most downloaded paper ever, in the Journal of Mathematical Psychology. In the paper, titled, Utility Independence of Multiattribute Utility Theory is Equivalent to Standard Sequence Invariance of Conjoint Measurement, the authors show that utility independence and standard sequences are closely related, and that utility independence is equivalent to a standard sequence invariance condition when applied to risk.
Multiattribute utility theory is widely used in business and healthcare as a means for analyzing risky decisions with multiple objectives such as the prospect of a new business venture or the value of a new medical treatment. At the center of multiattribute theory is an assumption called “utility independence” which holds that the preference for a particular objective in a decision is not affected by levels of another objective when that other objective is held constant. Utility independence is often needed to make the analysis of decision with multiple objectives tractable.
However, utility independence is difficult to test due to psychological biases. In contrast to Multiattribute theory, conjoint measurement is a different approach that measures value in the absence of risk using a technique involving the construction of standard sequences (similar to constructing equally spaced differences in value). This paper shows that utility independence and standard sequences are closely related: utility independence is equivalent to the invariance of standard sequences when applied to risk. This simple relation between two widely used conditions in adjacent fields of research is surprising and useful. It facilitates the testing of utility independence because standard sequences are flexible and can avoid cancellation biases that affect direct tests of utility independence. Thus a long held assumption in Multiattribute utility theory now has a means of testing to justify its use in business and health care applications.
Other paper authors include Han Bleichrodt, Martin Filko, and Peter P. Walker.